Human memory is notoriously inaccurate. Whereas newly acquired memories tend to have high validity, over time subsequent experiences can change or degrade them, a phenomenon termed "retroactive interference." This is not simple forgetting, but an active corruption of earlier memories by later ones. The central hypothesis of this project is that interference between memories results naturally from the brain's method of storing related information in partially overlapping fashion in neural networks. In this project we will study two memories that are stored together in a single network of the experimentally advantageous marine mollusk Tritonia diomedea. Using the tools of optical recording from neuronal populations with fast voltage-sensitive dyes, intracellular recording from individual neurons with sharp electrodes, and realistic network simulations of the animal's memory-storing network, we will test several cellular level hypotheses regarding the anatomical and functional organization of individual and multiple memories in a real brain. The project will utilize a new, hybrid microscope designed by us to facilitate the integration of conventional sharp electrode electrophysiology with large-scale optical recording of network activity. Our specific aims are: Aim 1. Expand our knowledge of Tritonia's memory-storing escape swim network, including the role of a newly-discovered class of "casually participating" neurons. Here we will characterize several new neurons discovered in our optical recordings, and test a set of hypotheses about network function. Aim 2. Map the first memory (sensitization). While memory has been intensively studied at the synaptic and molecular level, less is understood about its anatomical organization in nervous systems. How distributed are the cellular and synaptic changes encoding a single memory? How are the different components of information organized with respect to the distributed sites of plasticity underlying the memory? We will evaluate competing hypotheses regarding these issues by mapping out the memory for sensitization in the Tritonia brain. Aim 3. Map the second memory (habituation), and determine how it partially interferes with, and partially co-exists with the pre-existing memory for sensitization. Based on behavioral studies, habituation appears to interfere with some but not all components of an initial sensitization memory in Tritonia. Here we will map the memory for habituation as it develops, and will attempt to determine whether it does indeed erase the prior memory, or whether that memory persists, intact but hidden, in the same neural network. The long-term goal of this project is to better understand the network organization of memories in the brain, and how overlapping storage affects memory accuracy, with the goal of developing better treatments for conditions such as chronic anxiety, post-traumatic stress disorder, and memory dysfunction after brain injury. PUBLIC HEALTH RELEVANCE: This project investigates the degree to which behavioral functions and memories are organized in distributed, overlapping fashion in the brain. This important but poorly understood organizational scheme has relevance for phenomena ranging from mechanisms of recovery of brain function after injury, to retroactive interference, where later experiences so insidiously manage to alter our preexisting memories. This investigation is made possible by our development of a new technical approach for imaging large-scale network activity in a simple model preparation.